Hugging Face's logo

dyang39
/
InternVL-reasoning-2b-dozen

Safetensors
internvl_chat
custom_code
Model card Files
xet
Community
dyang39 commited on
Commit
68ca749
·
verified ·
1 Parent(s): c459ace

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +150 -0
trainer_state.json ADDED
@@ -0,0 +1,150 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 16.0,
5
+ "eval_steps": 500,
6
+ "global_step": 20,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.8,
13
+ "learning_rate": 4e-05,
14
+ "loss": 3.2717,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 1.6,
19
+ "learning_rate": 3.972722606805445e-05,
20
+ "loss": 3.1418,
21
+ "step": 2
22
+ },
23
+ {
24
+ "epoch": 2.4,
25
+ "learning_rate": 3.8916344834012695e-05,
26
+ "loss": 1.8546,
27
+ "step": 3
28
+ },
29
+ {
30
+ "epoch": 3.2,
31
+ "learning_rate": 3.758947502412978e-05,
32
+ "loss": 1.7453,
33
+ "step": 4
34
+ },
35
+ {
36
+ "epoch": 4.0,
37
+ "learning_rate": 3.578281018792788e-05,
38
+ "loss": 0.9852,
39
+ "step": 5
40
+ },
41
+ {
42
+ "epoch": 4.8,
43
+ "learning_rate": 3.354563143251483e-05,
44
+ "loss": 1.2799,
45
+ "step": 6
46
+ },
47
+ {
48
+ "epoch": 5.6,
49
+ "learning_rate": 3.093896316244855e-05,
50
+ "loss": 0.6405,
51
+ "step": 7
52
+ },
53
+ {
54
+ "epoch": 6.4,
55
+ "learning_rate": 2.8033908493059394e-05,
56
+ "loss": 0.5262,
57
+ "step": 8
58
+ },
59
+ {
60
+ "epoch": 7.2,
61
+ "learning_rate": 2.4909709742815986e-05,
62
+ "loss": 0.3255,
63
+ "step": 9
64
+ },
65
+ {
66
+ "epoch": 8.0,
67
+ "learning_rate": 2.165158690944665e-05,
68
+ "loss": 0.2122,
69
+ "step": 10
70
+ },
71
+ {
72
+ "epoch": 8.8,
73
+ "learning_rate": 1.8348413090553356e-05,
74
+ "loss": 0.1912,
75
+ "step": 11
76
+ },
77
+ {
78
+ "epoch": 9.6,
79
+ "learning_rate": 1.5090290257184019e-05,
80
+ "loss": 0.2122,
81
+ "step": 12
82
+ },
83
+ {
84
+ "epoch": 10.4,
85
+ "learning_rate": 1.1966091506940616e-05,
86
+ "loss": 0.1126,
87
+ "step": 13
88
+ },
89
+ {
90
+ "epoch": 11.2,
91
+ "learning_rate": 9.061036837551467e-06,
92
+ "loss": 0.0647,
93
+ "step": 14
94
+ },
95
+ {
96
+ "epoch": 12.0,
97
+ "learning_rate": 6.4543685674851834e-06,
98
+ "loss": 0.0087,
99
+ "step": 15
100
+ },
101
+ {
102
+ "epoch": 12.8,
103
+ "learning_rate": 4.217189812072131e-06,
104
+ "loss": 0.0087,
105
+ "step": 16
106
+ },
107
+ {
108
+ "epoch": 13.6,
109
+ "learning_rate": 2.410524975870221e-06,
110
+ "loss": 0.0036,
111
+ "step": 17
112
+ },
113
+ {
114
+ "epoch": 14.4,
115
+ "learning_rate": 1.0836551659873073e-06,
116
+ "loss": 0.0028,
117
+ "step": 18
118
+ },
119
+ {
120
+ "epoch": 15.2,
121
+ "learning_rate": 2.7277393194555357e-07,
122
+ "loss": 0.0016,
123
+ "step": 19
124
+ },
125
+ {
126
+ "epoch": 16.0,
127
+ "learning_rate": 0.0,
128
+ "loss": 0.0006,
129
+ "step": 20
130
+ },
131
+ {
132
+ "epoch": 16.0,
133
+ "step": 20,
134
+ "total_flos": 1.4917407062462824e+19,
135
+ "train_loss": 0.7294653365388513,
136
+ "train_runtime": 705.176,
137
+ "train_samples_per_second": 0.567,
138
+ "train_steps_per_second": 0.028
139
+ }
140
+ ],
141
+ "logging_steps": 1.0,
142
+ "max_steps": 20,
143
+ "num_input_tokens_seen": 0,
144
+ "num_train_epochs": 20,
145
+ "save_steps": 4,
146
+ "total_flos": 1.4917407062462824e+19,
147
+ "train_batch_size": 1,
148
+ "trial_name": null,
149
+ "trial_params": null
150
+ }